diff --git a/default.env b/default.env
index 6723905..34673cc 100644
--- a/default.env
+++ b/default.env
@@ -7,6 +7,9 @@ CACHE = 1
 N_THREADS = 4
 MAX_TOKENS = 1024
 
+# Bot specific settings
+REFRESH_INTERVAL=7
+
 # ROOT_IP is only used when running the bot without docker compose
 ROOT_IP = 192.168.0.15
 
diff --git a/llamabot.js b/llamabot.js
index dbffa96..efc0128 100644
--- a/llamabot.js
+++ b/llamabot.js
@@ -241,8 +241,8 @@ async function generateResponse(conversation, message) {
     // call the function initially
     await showSystemLoad();
 
-    // refresh the system load percentage and update the message every 7 seconds
-    const refreshInterval = setInterval(showSystemLoad, 7000);
+    // Grab the REFRESH_INTERVAL from ENV if not exist, lets use 7 (seconds)
+    const refreshInterval = setInterval(showSystemLoad, (process.env.REFRESH_INTERVAL || 7) * 1000);
 
     try {
         const response = await fetch(`http://${process.env.ROOT_IP}:${process.env.ROOT_PORT}/v1/chat/completions`, {